- 27 Oct, 2013 3 commits
-
-
Peng Tao authored
Running on 3.11-rc4 kernel, I got below endless loop. It turns to be that Lustre always saves the first page of a dir inode mapping at index ~0UL. And after commit 5a720394 (mm: teach truncate_inode_pages_range() to handle non page aligned ranges), truncate_inode_pages_range() _NO LONGER_ truncates the page that is sitting at index ~0UL. [16768.998006] mkdir R running task 0 2717 2716 0x00000080 [16768.998073] 000000000000000e 0000000000000000 0000000000000000 ffff88000be00460 [16768.998157] ffff88000ea65908 ffffffff810fec3e ffff88000ea65968 ffff8800229e7750 [16768.998241] ffff88000ea658b8 0000000000000000 0000000000000000 ffff88000ea65958 [16768.998326] Call Trace: [16768.998401] [<ffffffff810fc6ed>] ? rcu_read_unlock+0x1c/0x2d [16768.998473] [<ffffffff810fec3e>] ? find_get_pages+0xf5/0x11b [16768.998530] [<ffffffff811078f0>] ? pagevec_lookup+0x20/0x2a [16768.998586] [<ffffffff8110920e>] ? truncate_inode_pages_range.part.2+0x161/0x39a [16768.998680] [<ffffffffa02ad5dc>] ? ll_md_blocking_ast+0x338/0x62f [lustre] [16768.998744] [<ffffffff8110947f>] ? truncate_inode_pages_range+0x38/0x3f [16768.998805] [<ffffffff811094f8>] ? truncate_inode_pages+0x12/0x14 [16768.998871] [<ffffffffa02ad6e8>] ? ll_md_blocking_ast+0x444/0x62f [lustre] [16768.998948] [<ffffffff810981b5>] ? arch_local_irq_save+0x9/0xc [16768.999022] [<ffffffffa07ee0e8>] ? ldlm_cancel_callback+0x67/0x12a [ptlrpc] [16768.999100] [<ffffffffa07f85b2>] ? ldlm_cli_cancel_local+0xf3/0x2bc [ptlrpc] [16768.999176] [<ffffffffa07f9163>] ? ldlm_cli_cancel_list_local+0x7e/0x1e4 [ptlrpc] [16768.999268] [<ffffffffa07f9473>] ? ldlm_cancel_resource_local+0x1aa/0x1b9 [ptlrpc] [16768.999385] [<ffffffffa0657bad>] ? mdc_resource_get_unused+0xf8/0x115 [mdc] [16768.999472] [<ffffffff8109c887>] ? trace_hardirqs_on+0xd/0xf [16768.999533] [<ffffffffa06583d8>] ? mdc_create+0x11e/0x4db [mdc] [16768.999597] [<ffffffff8152ed84>] ? mutex_unlock+0xe/0x10 [16768.999654] [<ffffffffa0350e99>] ? lmv_create+0x355/0x3e9 [lmv] [16768.999712] [<ffffffff811553b7>] ? final_putname+0x35/0x39 [16768.999775] [<ffffffffa02ae167>] ? ll_new_node+0x33b/0x3ff [lustre] [16768.999841] [<ffffffffa02ae62c>] ? ll_mkdir+0xf2/0x127 [lustre] [16768.999897] [<ffffffff81156996>] ? vfs_mkdir+0x84/0xc9 [16768.999961] [<ffffffff81158cf8>] ? SyS_mkdirat+0x77/0xad [16769.000014] [<ffffffff81158d47>] ? SyS_mkdir+0x19/0x1b [16769.000066] [<ffffffff81538652>] ? system_call_fastpath+0x16/0x1b Signed-off-by:
Peng Tao <bergwolf@gmail.com> Signed-off-by:
Andreas Dilger <andreas.dilger@intel.com> Signed-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Peng Tao authored
It is left over when porting Lustre tree patch in commit (e62e5d92) and should be removed. Signed-off-by:
-
Lisa Nguyen authored
Remove typedef keyword and rename the cfs_hash_bucket_t struct to cfs_hash_bucket in libcfs_hash.h. These changes resolve the "Do not add new typedefs" warning generated by checkpatch.pl and meet kernel coding style. The struct variables in hash.c are updated to reflect this change as well. Signed-off-by:
Lisa Nguyen <lisa@xenapiadmin.com> Signed-off-by:
-
- 25 Oct, 2013 2 commits
-
-
Rashika Kheria authored
This patch fixes the following error in sb_pc_mp.c- drivers/staging/sb105x/sb_pci_mp.c:546 mp_startup() error: we previously assumed 'info->tty' could be null (see line 525) Signed-off-by:
Rashika Kheria <rashika.kheria@gmail.com> Reviewed-by:
Josh Triplett <josh@joshtriplett.org> Signed-off-by:
-
John Stultz authored
Hopefully this isn't too late for 3.12. In commit 7dc19d5a (convert shrinkers to new count/scan API) the return value to PURGE_ALL_CACHES was dropped, causing -EPERM to always be returned. This patch re-adds the ret assignment, setting it to the the ashmem_shrink_count(), which is the lru_count. (Sorry this was missed in the review!) Fixes: 7dc19d5a ("convert shrinkers to new count/scan API") Cc: Colin Cross <ccross@android.com> Cc: Android Kernel Team <kernel-team@android.com> Cc: Glauber Costa <glommer@openvz.org> Reported-by:
YongQin Liu <yongqin.liu@linaro.org> Signed-off-by:
John Stultz <john.stultz@linaro.org> Cc: stable <stable@kernel.org> # 3.12 Acked-by:
Dave Chinner <dchinner@redhat.com> Signed-off-by:
-
- 21 Oct, 2013 8 commits
-
-
Lisa Nguyen authored
Remove typedef keyword and rename the cfs_hash_lock_t union to cfs_hash_lock in libcfs_hash.h. These changes resolve the "Do not add new typedefs" warning generated by checkpatch.pl and meet kernel coding style. The union variables in hash.c are updated to reflect this change as well. Signed-off-by:
-
Dulshani Gunawardhana authored
Fix checkpatch.pl warning of "(foo*)" should be "(foo *)" in lustre/lov Signed-off-by:
Dulshani Gunawardhana <dulshani.gunawardhana89@gmail.com> Reviewed-by:
-
Dulshani Gunawardhana authored
Fix the checkpatch.pl warning 'externs should be avoided in .c files' in lustre/lov. Signed-off-by:
-
Dulshani Gunawardhana authored
Fix checkpatch.pl error "space required before the open parenthesis '('" in lustre/lov. Signed-off-by: -
Dulshani Gunawardhana authored
Replace max() with max_t() in accordance to chekpatch.pl warning in lov_obd.c. Signed-off-by:
-
Dulshani Gunawardhana authored
Fix C99 style comments in lov_obd.c in accordance to checkpatch.pl warning. Signed-off-by:
-
Dulshani Gunawardhana authored
Remove checkpatch.pl warning of space prohibited between function name and open parenthesis '('in /lustre/lov. Signed-off-by: -
Lisa Nguyen authored
The functions in the bitmap.h header file are not being used by any source file in the lustre driver. Unless there are any objections, please remove the bitmap.h header file as it contains unused code in the kernel. Signed-off-by:
-
- 19 Oct, 2013 27 commits
-
-
Teodora Baluta authored
Fix the following checkpatch.pl errors: drivers/staging/rtl8187se/ieee80211/ieee80211_softmac.c:890: "foo* bar" should be "foo *bar" drivers/staging/rtl8187se/ieee80211/ieee80211_softmac.c:944: "foo* bar" should be "foo *bar" Signed-off-by:
Teodora Baluta <teobaluta@gmail.com> Signed-off-by:
-
Teodora Baluta authored
This patch fixes sparse warnings regarding functions that should be static in driver rtl8187se. Signed-off-by:
-
Teodora Baluta authored
Fix the following type of sparse warnings: drivers/staging/rtl8187se/r8180_core.c:163:48: warning: incorrect type in argument 1 (different address spaces) drivers/staging/rtl8187se/r8180_core.c:163:48: expected void const volatile [noderef] <asn:2>*addr drivers/staging/rtl8187se/r8180_core.c:163:48: got unsigned char [usertype] * Signed-off-by:
-
Teodora Baluta authored
Fix the following type of sparse warnings: drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17: warning: incorrect type in argument 2 (different base types) drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17: expected restricted gfp_t [usertype] flags drivers/staging/rtl8187se/ieee80211/ieee80211_tx.c:249:17: got int [signed] gfp_mask Signed-off-by:
-
Rashika Kheria authored
This patch fixes the following Sparse information in sb_pci_mp.c- ~/git/kernels/linux/drivers/staging/sb105x/sb_pci_mp.c:1813 mp_unregister_driver() info: redundant null check on drv->state calling kfree() Signed-off-by:
-
Ashvini Varatharaj authored
drivers/staging/line6/driver.c:208 line6_send_raw_message_async_part() info: why not propagate 'retval' from usb_submit_urb() instead of (-22)? Signed-off-by:
Ashvini Varatharaj <ashvinivaratharaj@gmail.com> Reviewed-by:
-
Ashvini Varatharaj authored
drivers/staging/line6/midi.c:147 send_midi_async() info: why not propagate 'retval' from usb_submit_urb() instead of (-22)? Signed-off-by:
-
Ashvini Varatharaj authored
Fix checkpatch warning: WARNING: simple_strtoul is obsolete, use kstrtoul instead Signed-off-by:
Peter P Waskiewicz Jr <peter.p.waskiewicz.jr@intel.com> Signed-off-by:
-
Rashika Kheria authored
This patch fixes the following smatch warning- ~/git/kernels/linux/drivers/staging/sep/sep_crypto.c:1137 sep_crypto_block_data() info: why not propagate 'int_error' from sep_oddball_pages() instead of (-12)? ~/git/kernels/linux/drivers/staging/sep/sep_crypto.c:1152 sep_crypto_block_data() info: why not propagate 'int_error' from sep_oddball_pages() instead of (-12)? Signed-off-by:
-
Archana kumari authored
This patch fixes C99 commenting style in staging:vt6655:michael.h Signed-off-by:
Archana kumari <archanakumari959@gmail.com> Reviewed-by:
-
Rashika Kheria authored
This patch fixes the following checkpatch.pl warning in slicoss.c- WARNING: quoted string split across lines Signed-off-by:
-
Rashika Kheria authored
This patch fixes the following checkpatch.pl warning in slicoss.c: WARNING: Too many leading tabs - consider code refactoring Signed-off-by:
-
Rashika Kheria authored
This patch fixes the following checkpatch.pl warning in slicoss.c: WARNING: Prefer netdev_dbg(netdev, ... then dev_dbg(dev, ... then pr_debug(... to printk(KERN_DEBUG ... Signed-off-by:
-
Rashika Kheria authored
This patch fixes the following checkpatch.pl warning in slicoss.c: WARNING: debugfs_remove(NULL) is safe this check is probably not required Signed-off-by:
-
Ashvini Varatharaj authored
drivers/staging/dgnc/dgnc_neo.c:1969 neo_vpd() error: buffer overflow 'brd->vpd' 128 <= 128 drivers/staging/dgnc/dgnc_neo.c:1970 neo_vpd() warn: buffer overflow 'brd->vpd' 128 <= 130 drivers/staging/dgnc/dgnc_neo.c:1970 neo_vpd() warn: buffer overflow 'brd->vpd' 128 <= 130 Signed-off-by:
-
Ashvini Varatharaj authored
Fix dgnc_mgmt_ioctl() warn: unsigned 'board' is never less than zero. Signed-off-by:
-
Archana kumari authored
Fixes "braces {} are not necessary for single statement blocks" warning in dgnc_driver.c Signed-off-by: -
Archana kumari authored
Fixed removal of unecessary error messages in staging:dgnc:dgnc_driver.c Signed-off-by:
-
Archana kumari authored
This patch fixes sparse warning "Using plain integer as NULL pointer" in sta_h2.c Signed-off-by:
-
Archana kumari authored
This patch fixes the following warnings in hcf.c: "space prohibited between function name and open parenthesis '(' " "space prohibited before that close parenthesis ')' ". Signed-off-by: -
Ian Abbott authored
Some bits of the 'CRB' register have different functions when read or written, so add macros to define the read-only parts. Add macros to define the widths of the bitfields in the 'CRA' and 'CRB' registers and the standard encoder setup value. Add macros to construct and extract parts of the 'CRA' and 'CRB' register values and the standard encoder setup value, along with a couple of general helper macros for the above. Redefine the bitfield mask macros for 'CRA', 'CRB' and standard encoder setup using the above. Signed-off-by:
Ian Abbott <abbotti@mev.co.uk> Reviewed-by:
H Hartley Sweeten <hsweeten@visionengravers.com> Signed-off-by:
-
Ian Abbott authored
There are some bits in the 'CRB' register not defined in "s626.h". Three of these are read-only bits that overlay the write-only interrupt control bits. Another missing bit controls whether counter 'B' is cleared when counter 'A' overflows. Add the missing bit definitions for completeness. Signed-off-by:
-
Ian Abbott authored
The counter 'A' clock polarity field in the 'CRA' register is only 1 bit wide, but the `S626_CRAMSK_CLKPOL_A` macro shows it as 2 bits wide, which would overlap with the counter 'A' interrupt source field. This is harmless as the macro isn't actually used yet, but correct it anyway as I want to use it! Signed-off-by:
-
Ian Abbott authored
The macros `S626_CLKSRC_COUNTER`, `S626_CLKSRC_TIMER` and `S626_CLKSRC_EXTENDER` are used logically to set the operating mode of an encoder channel. `S626_CLKSRC_COUNTER` (0) is also used as a 2-bit physical value to set the counter source of an encoder channel to "encoder". Rename the macros to `S626_ENCMODE_COUNTER`, `S626_ENCMODE_TIMER` and `S626_ENCMODE_EXTENDER` and rename some other macros and (unused) functions relating to the encoder mode for consistency. Define new macros to specify the physical counter source values for the 'CRA' register and rename the corresponding bitshift and mask macros accordingly. The physical values for the counter source are: S626_CNTSRC_ENCODER = 0 // encoder S626_CNTSRC_DIGIN = 1 // digital inputs S626_CNTSRC_SYSCLK = 2 // system clock up S626_CNTSRC_SYSCLK_DOWN = 3 // system clock down Also use the `S626_CNTSRC_SYSCLK` value as a bitmask (bit 1) to indicate either of the system clock values, with the direction (bit 0) indicated separately in this case. Signed-off-by:
-
Ian Abbott authored
Signed-off-by:
-
Ian Abbott authored
When setting the clock source for one of the 'A' encoders to operate in "counter" mode in `s626_set_mode_a()`, bitshift the clock source value by `S626_CRABIT_CLKSRC_A` for consistency with the other modes. This has no effect on the value since `S626_CRABIT_CLKSRC_A` is 0. Signed-off-by:
-
Ian Abbott authored
When setting up an encoder channel, the setup value includes a polarity and direction, but these are the same bit of the setup value: S626_CLKPOL_POS = S626_CNTDIR_UP = 0 S626_CLKPOL_NEG = S626_CNTDIR_DOWN = 1 In the construction of the setup value, both the CLKPOL and the CNTDIR constants are shifted by the same amount `S626_BF_CLKPOL`. Only the following combinations are set up currently (this may change if user configuration of the encoder is implemented properly): (S626_CLKPOL_POS << S626_BF_CLKPOL) (S626_CLKPOL_POS << S626_BF_CLKPOL) | (S626_CNTDIR_UP << S626_BF_CLKPOL) (S626_CLKPOL_POS << S626_BF_CLKPOL) | (S626_CNTDIR_DOWN << S626_BF_CLKPOL) The first two are used in "counter" mode and is equivalent to: (S626_CLKPOL_POS << S626_BF_CLKPOL) The last one is used in "timer" mode and is equivalent to: (S626_CNTDIR_DOWN << S626_BF_CLKPOL) Use the shorter equivalents. The comments in "s626.h" indicate that the 'CLKPOL' constants make more sense for the "counter" mode (when the encoders operate as up/down counters) and the 'CNTDIR' constants make more sense for the "timer" mode. Signed-off-by:
-
